README.md

@protobufjs/float

Reads / writes floats / doubles from / to buffers in both modern and ancient browsers. Fast.

API

• writeFloatLE(val: number, buf: Uint8Array, pos: number)
  Writes a 32 bit float to a buffer using little endian byte order.

• writeFloatBE(val: number, buf: Uint8Array, pos: number)
  Writes a 32 bit float to a buffer using big endian byte order.

• readFloatLE(buf: Uint8Array, pos: number): number
  Reads a 32 bit float from a buffer using little endian byte order.

• readFloatBE(buf: Uint8Array, pos: number): number
  Reads a 32 bit float from a buffer using big endian byte order.

• writeDoubleLE(val: number, buf: Uint8Array, pos: number)
  Writes a 64 bit double to a buffer using little endian byte order.

• writeDoubleBE(val: number, buf: Uint8Array, pos: number)
  Writes a 64 bit double to a buffer using big endian byte order.

• readDoubleLE(buf: Uint8Array, pos: number): number
  Reads a 64 bit double from a buffer using little endian byte order.

• readDoubleBE(buf: Uint8Array, pos: number): number
  Reads a 64 bit double from a buffer using big endian byte order.

Performance

There is a simple benchmark included comparing raw read/write performance of this library (float), float's fallback for old browsers, the ieee754 module and node's buffer. On an i7-2600k running node 6.9.1 it yields:

benchmarking writeFloat performance ...

float x 42,741,625 ops/sec ±1.75% (81 runs sampled)
float (fallback) x 11,272,532 ops/sec ±1.12% (85 runs sampled)
ieee754 x 8,653,337 ops/sec ±1.18% (84 runs sampled)
buffer x 12,412,414 ops/sec ±1.41% (83 runs sampled)
buffer (noAssert) x 13,471,149 ops/sec ±1.09% (84 runs sampled)

                      float was fastest
           float (fallback) was 73.5% slower
                    ieee754 was 79.6% slower
                     buffer was 70.9% slower
          buffer (noAssert) was 68.3% slower

benchmarking readFloat performance ...

float x 44,382,729 ops/sec ±1.70% (84 runs sampled)
float (fallback) x 20,925,938 ops/sec ±0.86% (87 runs sampled)
ieee754 x 17,189,009 ops/sec ±1.01% (87 runs sampled)
buffer x 10,518,437 ops/sec ±1.04% (83 runs sampled)
buffer (noAssert) x 11,031,636 ops/sec ±1.15% (87 runs sampled)

                      float was fastest
           float (fallback) was 52.5% slower
                    ieee754 was 61.0% slower
                     buffer was 76.1% slower
          buffer (noAssert) was 75.0% slower

benchmarking writeDouble performance ...

float x 38,624,906 ops/sec ±0.93% (83 runs sampled)
float (fallback) x 10,457,811 ops/sec ±1.54% (85 runs sampled)
ieee754 x 7,681,130 ops/sec ±1.11% (83 runs sampled)
buffer x 12,657,876 ops/sec ±1.03% (83 runs sampled)
buffer (noAssert) x 13,372,795 ops/sec ±0.84% (85 runs sampled)

                      float was fastest
           float (fallback) was 73.1% slower
                    ieee754 was 80.1% slower
                     buffer was 67.3% slower
          buffer (noAssert) was 65.3% slower

benchmarking readDouble performance ...

float x 40,527,888 ops/sec ±1.05% (84 runs sampled)
float (fallback) x 18,696,480 ops/sec ±0.84% (86 runs sampled)
ieee754 x 14,074,028 ops/sec ±1.04% (87 runs sampled)
buffer x 10,092,367 ops/sec ±1.15% (84 runs sampled)
buffer (noAssert) x 10,623,793 ops/sec ±0.96% (84 runs sampled)

                      float was fastest
           float (fallback) was 53.8% slower
                    ieee754 was 65.3% slower
                     buffer was 75.1% slower
          buffer (noAssert) was 73.8% slower

To run it yourself:

$> npm run bench

License: BSD 3-Clause License